Service activity information processing method and device, electronic equipment and medium

Abstract

The invention relates to the technical field of data processing, artificial intelligence and the like, and discloses a business activity information processing method and device, electronic equipment and a medium. The business activity information processing method comprises the following steps: receiving multi-modal data fed back for a business activity site; performing semantic feature extraction on the multi-modal data to obtain semantic feature data of different modals; performing event aggregation processing on the semantic feature data to obtain a business activity event; and generating a service activity feedback result based on the service activity event. Through the method, valuable business activity information can be extracted from complex and huge information flow, and the processing efficiency of the business activity information is improved.

Description

Technical Field

[0001] This application relates to the fields of data processing, artificial intelligence and other technical fields, and in particular to a method, apparatus, electronic device and medium for processing business activity information. Background Technology

[0002] Large-scale business events, such as mid-year shopping mall celebrations, music festivals, and product launches, rely not only on planning but also on real-time management and rapid response capabilities during on-site execution. Managers need to have timely and comprehensive knowledge of on-site customer flow, material consumption, customer sentiment, and unexpected situations to make accurate resource allocation and emergency response decisions. Failure to collect and accurately process key on-site information not only results in a significant waste of valuable data assets but also increases the risk of errors during stressful events. Summary of the Invention

[0003] The embodiments of this application aim to at least partially address one of the technical problems in the related art. To this end, the embodiments of this application propose a method, apparatus, electronic device, and medium for processing business activity information.

[0004] The embodiments of this application provide a business activity information processing method, which includes: receiving multimodal data that provides feedback on the business activity site; extracting semantic features from the multimodal data to obtain semantic feature data of different modalities; performing event aggregation processing on the semantic feature data to obtain business activity events; and generating business activity feedback results based on the business activity events.

[0005] In some implementations, semantic feature data is subjected to event aggregation processing to obtain business activity events, including: aggregating multiple first information points in the semantic feature data to obtain business activity events, wherein the time interval between any two information points among the multiple first information points is less than a preset interval, and the topic similarity of the multiple first information points is greater than a preset similarity.

[0006] In some implementations, semantic feature data is subjected to event aggregation processing to obtain business activity events, including: in the process of event aggregation of semantic feature data, mutual verification and information enhancement processing of semantic feature data of different modalities are performed to obtain business activity events.

[0007] In some implementations, the method further includes: performing conflict event detection on multiple second information points in semantic feature data to obtain conflict detection results; and marking multiple second information points when the conflict detection results indicate that there is a conflict between multiple second information points.

[0008] In some implementations, a business activity event includes multiple business activity events; generating a business activity feedback result based on the business activity events includes: sorting the multiple business activity events according to time information and generating the business activity feedback result.

[0009] In some implementations, the method further includes: generating at least one of activity traffic data, activity problem hotspot data, and user sentiment data during the activity based on business activity feedback results and / or conflict detection results; and / or generating activity early warning information when the user sentiment data meets preset conditions.

[0010] In some implementations, the multimodal data includes at least one of raw text data, raw speech data, and raw visual data; semantic feature extraction of the multimodal data to obtain semantic feature data of different modalities includes: converting the raw speech data into text information with timestamps, and performing natural language understanding processing on the text information and the raw text data to obtain first semantic feature data, wherein the first semantic feature data includes at least one of key objects, core events, user sentiment tendencies, and activity intention classifications; and performing visual recognition on the raw visual data to obtain second semantic feature data, wherein the second semantic feature data includes at least one of key entities, activity scenarios, and user emotions.

[0011] The embodiments of this application provide a business activity information processing apparatus, which includes: a receiving module for receiving multimodal data that provides feedback on a business activity site; an extraction module for extracting semantic features from the multimodal data to obtain semantic feature data of different modalities; an aggregation module for performing event aggregation processing on the semantic feature data to obtain business activity events; and a generation module for generating business activity feedback results based on the business activity events.

[0012] An embodiment of this application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by one or more processors, which are executed by one or more processors to cause the one or more processors to implement the steps of the method of any of the above embodiments.

[0013] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0014] The business activity information processing method provided in this application can extract semantic feature data from different modalities and perform event aggregation analysis, thereby extracting event-centric business activity events from complex and fragmented feedback data. The final business activity feedback results not only clearly show the overall evolution of the activity, but also provide intuitive and reliable data support for activity review, problem identification, and process optimization, significantly improving the accuracy and efficiency of management decisions and maximizing the value of data. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a business activity information processing method provided for an embodiment of this application; Figure 2 A schematic diagram illustrating the process of real-time aggregation of multi-source, multi-modal information provided for the implementation of this application; Figure 3 A flowchart illustrating the event aggregation processing and conflict detection provided for the implementation of this application; Figure 4 A flowchart illustrating the process of generating business insights, visualizing them, and proactively issuing warnings for implementation of this application; Figure 5 A schematic diagram illustrating the process of extracting semantic feature data of different modalities for embodiments of this application; Figure 6 A flowchart illustrating the business activity information processing method provided for embodiments of this application; Figure 7 A schematic diagram of a business activity information processing device provided for an embodiment of this application; Figure 8 A block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0016] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0017] For large-scale business events such as shopping mall mid-year celebrations, music festivals, and product launches, success depends not only on planning but also on real-time management and rapid response capabilities during on-site execution. Managers need to have timely and comprehensive knowledge of on-site customer flow dynamics, material consumption, customer sentiment, and unexpected situations in order to make accurate resource allocation and emergency response decisions.

[0018] Information generated at an event is typically produced by frontline employees in various roles at different times and locations, in fragmented forms including voice, text, and images. Managers often only receive fragmented information, like "blind men touching an elephant." Traditional technologies are unable to intelligently aggregate these multi-source, heterogeneous, and fragmented information streams to reconstruct a complete picture of the event.

[0019] Moreover, the evolution of a problem on-site often exhibits continuity in both time and space. For example, employee A reports "long queue" in area A at 3:00 PM, and employee B also reports "long queue" in area B at 3:30 PM. Traditional technologies cannot automatically connect these two isolated points of information and identify the trend risk that "the queue problem is spreading."

[0020] Even if all feedback is collected, it's still just a pile of raw data. Managers need to spend a lot of effort reading, understanding, and relating it to extract valuable business insights. Traditional technologies lack an intelligent analytics engine that can automatically extract key events, identify group sentiments, and generate business insights that can guide decision-making from the integrated information.

[0021] Currently, on-site communication primarily relies on instant messaging tools (such as online work groups) and traditional walkie-talkies. In online work groups, discussions on different topics, reports of varying urgency, and casual conversations among different individuals intertwine, creating a chaotic and disorganized information flow. Key information is easily buried, making it difficult for managers to access it efficiently.

[0022] Furthermore, the voice messages, images, and text messages within the work group are unstructured, making direct data analysis impossible. After the event, reviewing the proceedings to identify the most frequent customer complaints becomes virtually impossible, resulting in a significant waste of valuable data assets. Additionally, each employee can only report from their own perspective, lacking a holistic view. Managers are essentially intelligence receivers, receiving fragmented information and needing to piece it together in a complex mental process—extremely difficult and error-prone in the stressful environment of a live event.

[0023] In view of this, this application provides a business activity information processing method that can automatically collect relevant information at the activity site and intelligently analyze the collected information to obtain on-site situational awareness results.

[0024] Figure 1 This is a flowchart illustrating a business activity information processing method provided for an embodiment of this application.

[0025] like Figure 1 As shown, the business activity information processing method 100 provided in this application includes, for example, steps S110-S140.

[0026] Step S110: Receive multimodal data that provides feedback on business activities at the site.

[0027] Multimodal data refers to feedback from on-site business activities, such as information provided by on-site staff, including security personnel, material handlers, and customer service personnel. Multimodal data includes feedback in the form of voice, text, images, and short videos.

[0028] Step S120: Extract semantic features from the multimodal data to obtain semantic feature data for different modalities.

[0029] A unified, deep semantic feature extraction process is performed on multimodal data to obtain semantic feature data for different modalities. Semantic feature data can be, for example, standardized semantic feature vectors that can be used for machine association analysis.

[0030] Step S130: Perform event aggregation processing on the semantic feature data to obtain business activity events.

[0031] For example, by aggregating and analyzing semantic feature data from different sources, centered on events, business activity events can be obtained. These business activity events may be candidate events or conflicting events containing contradictory information, formed based on the content of the aggregation analysis.

[0032] Step S140: Generate business activity feedback results based on business activity events.

[0033] For example, the feedback results of business activities can be generated by arranging related business activity events in chronological order to create a dynamic and traceable "event timeline" of the event, which can completely reconstruct the evolution of the event.

[0034] As can be seen, the business activity information processing method provided in this application can extract semantic feature data from different modalities and perform event aggregation analysis, thereby extracting business activity events centered on events from complex and fragmented feedback data. The final business activity feedback results not only clearly show the overall evolution of the activity, but also provide intuitive and reliable data support for activity review, problem identification, and process optimization, significantly improving the accuracy and efficiency of management decisions and maximizing the value of data.

[0035] Figure 2 This is a schematic diagram illustrating the process of real-time aggregation of multi-source, multi-modal information provided for the implementation of this application.

[0036] like Figure 2 As shown, the process of real-time aggregation of multi-source and multi-modal information includes steps S210-S220.

[0037] In step S210, the system can provide a unified information upload portal (such as a mobile app or mini-program) for all on-site personnel, such as security personnel, material handlers, and customer service personnel, to provide real-time feedback of multi-source, multi-modal information. Users can input on-site information at any time via voice, text, images, short videos, etc.

[0038] In step S220, when the system receives each message, it automatically marks it with rich metadata, including unique identifier, feedback person, feedback person role, feedback time, geographical location, etc.

[0039] Figure 3 This is a flowchart illustrating the event aggregation processing and conflict detection provided for the implementation of this application.

[0040] In one example, semantic feature data is subjected to event aggregation processing to obtain business activity events, including: aggregating multiple first information points in the semantic feature data to obtain business activity events, wherein the time interval between any two information points among the multiple first information points is less than a preset interval, and the topic similarity of the multiple first information points is greater than a preset similarity.

[0041] like Figure 3 As shown, multiple first information points in the semantic feature data are, for example, information points in the "spatiotemporal clustering" style of the semantic feature data. The preset interval is, for example, 5 minutes. The time interval between any two information points among the multiple first information points is less than the preset interval. For example, the system can automatically aggregate information points with similar times, such as within 5 minutes, to form a candidate event.

[0042] In semantic feature data, multiple primary information points may all contain entity information such as "gift" or "out of stock." If the similarity of the themes of these multiple primary information points is greater than a preset similarity, information points containing entity information such as "gift" or "out of stock" can be automatically aggregated to form a candidate event. For example, user A's feedback at 3 o'clock that "gift A seems to be almost gone" and user B's photo uploaded at 3:05 that "gift A shelf is empty" can be aggregated into a candidate event "gift A is out of stock." This candidate event can be directly used as a business activity event, or it can be verified, and the candidate event without conflicting information can be used as the business activity event.

[0043] In one example, semantic feature data is subjected to event aggregation processing to obtain business activity events. This includes: in the process of event aggregation of semantic feature data, mutual verification and information enhancement processing of semantic feature data of different modalities are performed to obtain business activity events.

[0044] like Figure 3As shown, during the event aggregation process of semantic feature data, cross-validation and information enhancement processing can be performed on semantic feature data of different modalities. For example, text descriptions can provide context for images, and images can provide objective evidence for text. Through cross-validation and information enhancement processing, the credibility and richness of business activity events can be enhanced.

[0045] In one example, the method further includes: performing conflict event detection on multiple second information points in semantic feature data to obtain conflict detection results; and marking multiple second information points when the conflict detection results indicate that there is a conflict between multiple second information points.

[0046] like Figure 3 As shown, when the system identifies a conflict between multiple secondary information points, for example, user A reports "sparse customer flow" in one area, while user B reports "crowded" in the same area 5 minutes later, the system will mark this as "conflicting information" and prompt the administrator to verify it further.

[0047] As can be seen, the embodiments provided in this application can automatically identify and aggregate independent and complete business activity events from scattered semantic features. Through spatiotemporal correlation analysis, the system can, like a human expert, discover deep connections between isolated information points and identify the evolution trend and potential risks of problems. During the aggregation process, semantic feature data from different modalities can be comprehensively utilized. Through cross-modal information mutual verification and enhancement, the credibility and richness of business activity events can be improved. Furthermore, conflict event detection can be performed on information points. Information points indicating conflict in the conflict detection results can be marked, providing clear guidance for subsequent manual review, debriefing analysis, and other stages. This ensures the accuracy and credibility of the subsequently generated business activity feedback results, becoming a valuable data asset that effectively guides future activity planning and maximizes the value of the data.

[0048] Figure 4 A schematic diagram illustrating the process of generating business insights, visualizing them, and proactively issuing warnings for implementation methods of this application.

[0049] In one example, a business activity event includes multiple business activity events; based on the business activity events, a business activity feedback result is generated, including: sorting the multiple business activity events according to time information and generating the business activity feedback result.

[0050] like Figure 4 As shown, the system sorts all the associated and confirmed events according to time information, generating a dynamic and traceable business activity feedback result. This business activity feedback result can reflect the "event timeline" of the event site and can completely reconstruct the evolution process of the event.

[0051] In one example, the business activity information processing method further includes: generating at least one of activity traffic data, activity problem hotspot data, and user sentiment data during the activity based on business activity feedback results and / or conflict detection results; and / or generating activity early warning information when the user sentiment data meets preset conditions.

[0052] like Figure 4 As shown, the business activity information processing method provided in this application can automatically perform statistics and insight mining based on on-site situation reconstruction and AI analysis engine. For example, by statistically analyzing the frequency and density of information containing "crowds" and "queues," a real-time heat map and time change curve of activity passenger flow can be drawn. By statistically analyzing the most frequent keywords in negative feedback, the "three major complaints of this activity" can be automatically identified. Furthermore, the distribution of positive and negative emotional feedback over time can be analyzed to generate a "user sentiment barometer during the activity."

[0053] like Figure 4 As shown in the embodiments provided in this application, the generated business activity feedback results can be ultimately presented in a visual manner on the manager's command screen or mobile terminal, such as dynamic maps, trend charts, keyword clouds, etc. Furthermore, the system can be configured with early warning rules; for example, when "negative sentiment" feedback exceeds a threshold within 10 minutes, an early warning notification is automatically sent to the person in charge.

[0054] Figure 5 This is a schematic diagram illustrating the process of extracting semantic feature data of different modalities for embodiments of this application.

[0055] In one example, the multimodal data includes at least one of raw text data, raw speech data, and raw visual data; semantic feature extraction is performed on the multimodal data to obtain semantic feature data of different modalities, including: converting the raw speech data into text information with timestamps, and performing natural language understanding processing on the text information and the raw text data to obtain first semantic feature data, wherein the first semantic feature data includes at least one of key objects, core events, user sentiment tendencies, and activity intention classifications; and performing visual recognition on the raw visual data to obtain second semantic feature data, wherein the second semantic feature data includes at least one of key entities, activity scenarios, and user emotions.

[0056] like Figure 5As shown, a unified and in-depth semantic feature extraction is performed on the real-time information stream gathered in the background. Specifically, the raw voice data is converted into text information with timestamps. All text information is processed to extract key objects (such as gift A, main stage, etc.), core events (such as start of redemption, out of stock, etc.), user sentiment (positive, negative, neutral) and activity intent classification (such as problem feedback, status synchronization, material request, etc.).

[0057] The original visual data includes images and videos. The images and videos are identified based on the CV model (Computer Vision Model) to obtain second semantic feature data. The second semantic feature data includes at least one of key entities, activity scenes, and user emotions. Key entities include crowds, long lines, empty shelves, etc., and user emotions include joyful facial expressions, etc.

[0058] like Figure 5 As shown, the output of the above steps can be a standardized semantic feature vector that can be used by machines for association analysis.

[0059] Figure 6 A flowchart illustrating the business activity information processing method provided in this application embodiment.

[0060] Figure 6 This diagram shows a complete flowchart of the business activity information processing method, the details of which are the same as those described above. Figures 2-5 The content described is the same or similar, so it will not be repeated here.

[0061] Figure 7 This is a schematic diagram of a business activity information processing device provided for an embodiment of this application.

[0062] like Figure 7 As shown, the business activity information processing device 700 includes: The receiving module 710 is used to receive multimodal data that provides feedback on business activities.

[0063] The extraction module 720 is used to extract semantic features from multimodal data to obtain semantic feature data of different modalities.

[0064] The aggregation module 730 is used to perform event aggregation processing on semantic feature data to obtain business activity events.

[0065] The generation module 740 is used to generate business activity feedback results based on business activity events.

[0066] For example, the aggregation module 730 is further configured to: aggregate multiple first information points in semantic feature data to obtain business activity events, wherein the time interval between any two information points among the multiple first information points is less than a preset interval, and the topic similarity of the multiple first information points is greater than a preset similarity.

[0067] For example, the aggregation module 730 is also used to: perform mutual verification and information enhancement processing on semantic feature data of different modalities during the event aggregation process of semantic feature data to obtain business activity events.

[0068] For example, the business activity information processing device 700 further includes a detection module, which is used to: perform conflict event detection on multiple second information points in semantic feature data to obtain conflict detection results; and mark multiple second information points when the conflict detection results indicate that there is a conflict between multiple second information points.

[0069] For example, the business activity event includes multiple business activity events; the generation module 740 is further configured to: sort the multiple business activity events according to time information and generate business activity feedback results.

[0070] For example, the generation module 740 is further configured to: generate at least one of activity traffic data, activity problem hotspot data, and user sentiment data during the activity based on business activity feedback results and / or conflict detection results; and / or generate activity warning information if the user sentiment data meets preset conditions.

[0071] For example, the multimodal data includes at least one of raw text data, raw speech data, and raw visual data; the extraction module 720 is further configured to: convert the raw speech data into text information with a timestamp, and perform natural language understanding processing on the text information and the raw text data to obtain first semantic feature data, wherein the first semantic feature data includes at least one of key objects, core events, user sentiment tendencies, and activity intention classifications; and perform visual recognition on the raw visual data to obtain second semantic feature data, wherein the second semantic feature data includes at least one of key entities, activity scenarios, and user emotions.

[0072] It is understood that the specific functions of the business activity information processing device 700 can be referred to the business activity information processing method above, and will not be repeated here.

[0073] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0074] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0075] This application provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0076] Figure 8 A block diagram of an electronic device provided in an embodiment of this application.

[0077] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method in any of the above embodiments.

[0078] like Figure 8 As shown, for ease of understanding, embodiments of this application illustrate a specific electronic device 800.

[0079] Electronic device 800 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0080] like Figure 8 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of electronic device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0081] Multiple components in electronic device 800 are connected to I / O interface 805. These components include: input unit 806, such as a keyboard or mouse; output unit 807, such as various types of displays or speakers; storage unit 808, such as a disk or optical disk; and communication unit 809, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 809 allows electronic device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0082] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0083] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0084] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0085] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0086] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0087] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0088] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0089] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for processing business activity information, characterized in that, The method includes: Receive multimodal data as feedback from on-site business activities; Semantic features are extracted from the multimodal data to obtain semantic feature data for different modalities; The semantic feature data is subjected to event aggregation processing to obtain business activity events; Based on the aforementioned business activity events, business activity feedback results are generated.

2. The method according to claim 1, characterized in that, The process of aggregating the semantic feature data to obtain business activity events includes: The business activity event is obtained by aggregating multiple first information points in the semantic feature data, wherein the time interval between any two information points is less than a preset interval, and the topic similarity of the multiple first information points is greater than a preset similarity.

3. The method according to claim 1, characterized in that, The process of aggregating the semantic feature data to obtain business activity events includes: During the event aggregation process of the semantic feature data, the semantic feature data of different modalities are mutually verified and information enhancement processes are performed to obtain the business activity events.

4. The method according to claim 1, characterized in that, Also includes: Conflict event detection is performed on multiple second information points in the semantic feature data to obtain conflict detection results; If the conflict detection result indicates that there is a conflict among the plurality of second information points, the plurality of second information points are marked.

5. The method according to claim 1, characterized in that, The business activity events include multiple business activity events; The step of generating business activity feedback results based on the business activity event includes: The multiple business activity events are sorted according to time information to generate the business activity feedback results.

6. The method according to claim 4, characterized in that, Also includes: Based on the business activity feedback results and / or the conflict detection results, generate at least one of the following: activity traffic data, activity problem hotspot data, and user sentiment data during the activity; and / or When the user's emotional data meets preset conditions, an activity warning message is generated.

7. The method according to any one of claims 1-6, characterized in that, The multimodal data includes at least one of raw text data, raw speech data, and raw visual data; the extraction of semantic features from the multimodal data to obtain semantic feature data of different modalities includes: The original voice data is converted into text information with timestamps, and natural language understanding processing is performed on the text information and the original text data to obtain first semantic feature data, wherein the first semantic feature data includes at least one of key objects, core events, user sentiment tendencies and activity intention categories; Visual recognition is performed on the original visual data to obtain second semantic feature data, wherein the second semantic feature data includes at least one of key entities, activity scenes, and user emotions.

8. A business activity information processing device, characterized in that, The device includes: The receiving module is used to receive multimodal data that is fed back from the business activity site; The extraction module is used to extract semantic features from the multimodal data to obtain semantic feature data of different modalities; The aggregation module is used to perform event aggregation processing on the semantic feature data to obtain business activity events; The generation module is used to generate business activity feedback results based on the business activity events.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images